| 1 |
On Mon, Jan 08, 2018 at 08:44:04PM -0600, R0b0t1 wrote |
| 2 |
|
| 3 |
> I was under the impression that disabling SMP on single core systems |
| 4 |
> could lead to a performance increase, but wasn't necessary. |
| 5 |
|
| 6 |
Quote from the "Help" in "make menuconfig" from the CONFIG_SMP item: |
| 7 |
|
| 8 |
==================================================================== |
| 9 |
This enables support for systems with more than one CPU. If you have |
| 10 |
a system with only one CPU, say N. If you have a system with more |
| 11 |
than one CPU, say Y. |
| 12 |
|
| 13 |
If you say N here, the kernel will run on uni- and multiprocessor |
| 14 |
machines, but will use only one CPU of a multiprocessor machine. If |
| 15 |
you say Y here, the kernel will run on many, but not all, |
| 16 |
uniprocessor machines. On a uniprocessor machine, the kernel |
| 17 |
will run faster if you say N here. |
| 18 |
==================================================================== |
| 19 |
|
| 20 |
> Is the deciding factor actually processor family? And - there are GCC |
| 21 |
> versions or distributions that have already dropped support for x86_64 |
| 22 |
> processors? I'd only read about 32 bit families being dropped. |
| 23 |
|
| 24 |
I think you're misunderstanding me. This is not about dropping x86_64 |
| 25 |
support. Rather, the earliest "Athlon" models came out *BEFORE* AMD |
| 26 |
introduced 64-bit support, and therefore support only 32-bit mode. |
| 27 |
Similarly Intel Pentium3 cpus support only 32-bit mode. |
| 28 |
|
| 29 |
-- |
| 30 |
Walter Dnes <waltdnes@××××××××.org> |
| 31 |
I don't run "desktop environments"; I run useful applications |
Archives